Combining machine learning and molecular simulations to predict the stability of amorphous drugs.

Amorphous drugs represent an intriguing option to bypass the low solubility of many crystalline formulations of pharmaceuticals. The physical stability of the amorphous phase with respect to the crystal is crucial to bring amorphous formulations into the market-however, predicting the timescale involved with the onset of crystallization a priori is a formidably challenging task. Machine learning can help in this context by crafting models capable of predicting the physical stability of any given amorphous drug. In this work, we leverage the outcomes of molecular dynamics simulations to further the state-of-the-art. In particular, we devise, compute, and use "solid state" descriptors that capture the dynamical properties of the amorphous phases, thus complementing the picture offered by the "traditional," "one-molecule" descriptors used in most quantitative structure-activity relationship models. The results in terms of accuracy are very encouraging and demonstrate the added value of using molecular simulations as a tool to enrich the traditional machine learning paradigm for drug design and discovery.

Understanding the emergence of the boson peak in molecular glasses.

A common feature of glasses is the "boson peak", observed as an excess in the heat capacity over the crystal or as an additional peak in the terahertz vibrational spectrum. The microscopic origins of this peak are not well understood; the emergence of locally ordered structures has been put forward as a possible candidate. Here, we show that depolarised Raman scattering in liquids consisting of highly symmetric molecules can be used to isolate the boson peak, allowing its detailed observation from the liquid into the glass. The boson peak in the vibrational spectrum matches the excess heat capacity. As the boson peak intensifies on cooling, wide-angle x-ray scattering shows the simultaneous appearance of a pre-peak due to molecular clusters consisting of circa 20 molecules. Atomistic molecular dynamics simulations indicate that these are caused by over-coordinated molecules. These findings represent an essential step toward our understanding of the physics of vitrification.